Reliability and validity for three autism instruments were compared for 190 children with low functioning autism (LFA), 190 children with high functioning autism or Asperger's disorder (HFA), 76 children with attention deficit hyperactivity disorder (ADHD), and 64 typical children. The instruments were the Checklist for Autism Spectrum Disorder (designed for children with LFA and HFA), Childhood Autism Rating Scale (CARS) for children with LFA, and Gilliam Asperger's Disorder Scale (GADS). For children with LFA or ADHD, classification accuracy was 100% for the Checklist and 98% for the CARS clinician scores. For children with HFA or ADHD, classification accuracy was 99% for the Checklist and 93% for the GADS clinician scores. Clinician-parent diagnostic agreement was high (90% Checklist, 90% CARS, and 84% GADS).

With the increase of students with autism in public schools, alternative schools, and in-home programming, attention to effective learning procedures has increased. A general model based on precision teaching for practicing behavior to fluency is presented as a guide for teachers and parents. This model applies to a wide variety of skills and can be tailored for individual students. The procedure for conducting efficient practice fall into four broad categories: (a) planning the practice routine, (b) developing the practice routine, (c) implementing the practice methods, and (d) analysis of data.

The authors of the Childhood Autism Rating Scale (CARS) state in the manual that the best cutoff score for distinguishing low functioning autism (LFA) from intellectual disability is 30 for children and 28 for adolescents and adults. This study determined that a cutoff score of 25.5 was most accurate in differentiating between high functioning autism or Asperger syndrome (HFA; n = 197) and ADHD (n = 74) in a sample of 1- to 16-year-olds with IQs of 80 or higher. Classification accuracy was 96% using clinician scores and 72% using parent scores. Children with LFA (n = 193) had significantly higher clinician and parent scores than children with HFA, and scores were negatively correlated with IQ. None of the typical children (n = 64) earned parent scores greater than 21.

Little is known about the validity of Gilliam Asperger's Disorder Scale (GADS), although it is widely used. This study of 199 children with high functioning autism or Asperger's disorder, 195 with low functioning autism, and 83 with attention deficit hyperactivity disorder (ADHD) showed high classification accuracy (autism vs. ADHD) for clinicians' GADS Quotients (92%), and somewhat lower accuracy (77%) for parents' Quotients. Both children with high and low functioning autism had clinicians' Quotients (M=99 and 101, respectively) similar to the Asperger's Disorder mean of 100 for the GADS normative sample. Children with high functioning autism scored significantly higher on the cognitive patterns subscale than children with low functioning autism, and the latter had higher scores on the remaining subscales: social interaction, restricted patterns of behavior, and pragmatic skills. Using the clinicians' Quotient and Cognitive Patterns score, 70% of children were correctly identified as having high or low functioning autism or ADHD.